Teven

pet. This repository contains the code for "How many data points is a prompt worth?"

BasicSR. Basic Super-Resolution Toolbox, including SRResNet, SRGAN, ESRGAN, etc.

sentence-transformers. Multilingual Sentence & Image Embeddings with BERT

deduplicate-text-datasets. Rust

Megatron-DeepSpeed. Ongoing research training transformer language models at scale, including: BERT & GPT-2

mpww. Code to recreate the MPWW dataset

glucose. GLUCOSE: GeneraLized and COntextualized Story Explanations https://arxiv.org/abs/2009.07758